Limits...
The human otubain2-ubiquitin structure provides insights into the cleavage specificity of poly-ubiquitin-linkages.

Altun M, Walter TS, Kramer HB, Herr P, Iphöfer A, Boström J, David Y, Komsany A, Ternette N, Navon A, Stuart DI, Ren J, Kessler BM - PLoS ONE (2015)

Bottom Line: Here we report the crystal structure of human otubain 2 (OTUB2) in complex with a ubiquitin-based covalent inhibitor, Ub-Br2.The ubiquitin binding mode is oriented differently to how viral otubains (vOTUs) bind ubiquitin/ISG15, and more similar to yeast and mammalian OTUs.In contrast to OTUB1 which has exclusive specificity towards Lys48 poly-ubiquitin chains, OTUB2 cleaves different poly-Ub linked chains.

View Article: PubMed Central - PubMed

Affiliation: Target Discovery Institute, Nuffield Department of Medicine, Roosevelt Drive, University of Oxford, Oxford, OX3 7FZ, United Kingdom; Science for Life Laboratory, Division of Translational Medicine and Chemical Biology, Department of Medical Biochemistry and Biophysics, Karolinska Institutet, SE-171 21 Stockholm, Sweden.

ABSTRACT
Ovarian tumor domain containing proteases cleave ubiquitin (Ub) and ubiquitin-like polypeptides from proteins. Here we report the crystal structure of human otubain 2 (OTUB2) in complex with a ubiquitin-based covalent inhibitor, Ub-Br2. The ubiquitin binding mode is oriented differently to how viral otubains (vOTUs) bind ubiquitin/ISG15, and more similar to yeast and mammalian OTUs. In contrast to OTUB1 which has exclusive specificity towards Lys48 poly-ubiquitin chains, OTUB2 cleaves different poly-Ub linked chains. N-terminal tail swapping experiments between OTUB1 and OTUB2 revealed how the N-terminal structural motifs in OTUB1 contribute to modulating enzyme activity and Ub-chain selectivity, a trait not observed in OTUB2, supporting the notion that OTUB2 may affect a different spectrum of substrates in Ub-dependent pathways.

Show MeSH

Related in: MedlinePlus

OTUB2 has a broader cleavage profile than OTUB1.(A) Ubiquitin (Ub), Nedd8, ISG15 and SUMO1 were conjugated to the biotinylated peptide VKAKIQD (Ub26–32) as described in [19] and subjected to cleavage by recombinant OTUB2, UCH-L3 or crude cell lysate (- represents untreated control), followed by SDS-PAGE separation and analysis by streptavidin-HRP immunoblotting. (B) di-SUMO2/3 was incubated with DMSO, OTUB2, OTUB2delta, OTUB1 or crude lysate for the indicated times, followed by SDS-PAGE separation and analysis by immunoblotting. (C) Linear di-Ubiquitin (di-Ub) was incubated with OTUB2, UCH-L3 or DMSO for the indicated times, followed by SDS-PAGE separation and analysis by immunoblotting. (D) di-Ub substrates with the linkages Lys6, 11, 27, 29, 33, 48 or 63 were incubated with either wild type (Wt) or catalytically inactive C51S mutant (M) OTUB2 for four hours, followed by SDS-PAGE and immunoblotting analysis.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4295869&req=5

pone.0115344.g005: OTUB2 has a broader cleavage profile than OTUB1.(A) Ubiquitin (Ub), Nedd8, ISG15 and SUMO1 were conjugated to the biotinylated peptide VKAKIQD (Ub26–32) as described in [19] and subjected to cleavage by recombinant OTUB2, UCH-L3 or crude cell lysate (- represents untreated control), followed by SDS-PAGE separation and analysis by streptavidin-HRP immunoblotting. (B) di-SUMO2/3 was incubated with DMSO, OTUB2, OTUB2delta, OTUB1 or crude lysate for the indicated times, followed by SDS-PAGE separation and analysis by immunoblotting. (C) Linear di-Ubiquitin (di-Ub) was incubated with OTUB2, UCH-L3 or DMSO for the indicated times, followed by SDS-PAGE separation and analysis by immunoblotting. (D) di-Ub substrates with the linkages Lys6, 11, 27, 29, 33, 48 or 63 were incubated with either wild type (Wt) or catalytically inactive C51S mutant (M) OTUB2 for four hours, followed by SDS-PAGE and immunoblotting analysis.

Mentions: We have searched for evidence for regulation of OTUB2 enzymatic activity. As shown previously, OTUB2 cleaved a Ub-based peptide substrate harbouring an isopeptide bond (Fig. 5A). Interestingly, we also noted cross-reactivity towards cleaving a NEDD8-based peptide substrate (Fig. 5A), although this may be a substrate-specific trait [18]. OTUB2 did not show any activity towards the ISG15-based peptide substrate, SUMO1, 2 or 3 (Fig. 5A and B) nor linear di-Ub (Fig. 5C). In contrast to OTUB1 which has exclusive specificity towards Lys48-linked chains [34], OTUB2 cleaves a broader range of di-Ub linked by naturally occurring isopeptide linkages with a preference for Lys63 di-Ub (Fig. 5D), consistent with previous studies [18]. A short C-terminal truncation (1–229; OTUB2ΔC5) did not markedly affect activity (S1A and C Fig., [5]), and no post-translational modifications within the protein were detected (data not shown). OTUB1’s strict selectivity towards cleaving Lys48-linked poly-Ub chains is in part due to its N-terminal properties [14, 15, 18]. OTUB2 has a shorter N-terminal tail and therefore might lack this feature to control for cleavage specificity. To test this hypothesis, we prepared chimeric constructs where the N-terminal tails of OTUB1 and OTUB2 were swapped to create N-term OTUB1-OTUB2 (OTUB1–2) and N-term OTUB2-OTUB1 (OTUB2–1) recombinant proteins (Fig. 6A). The OTUB1 N-terminal tails (1–83) and OTUB2 (1–43) were designated such that the OTU domain was left intact (S2 Fig.). Interestingly, active site labeling with either Br2 or VME based ubiquitin probes indicated that the OTUB1 N-terminal tail affects labeling selectivity of OTUB2 (OTUB1–2) towards the VME probe (Fig. 6B). Furthermore, OTUB2 enzymatic activity was restricted due to the presence of the OTUB1 N-terminal tail, and OTUB1 activity was enhanced in the presence of the OTUB2 N-terminal tail (OTUB2–1, Fig. 6C, D and E). Consistent with this, we observed that the presence of the OTUB1-N-terminal tail on OTUB2 (OTUB1–2) influenced its selectivity to cleave Lys63-tetra-ubiquitin chains when wild type and chimera OTUB1&2 recombinant proteins were subjected to a tetra-ubiquitin cleavage assay (Fig. 6F and G). Notably, the exclusive selectivity of OTUB1 for Lys48-linked di/tetra-ubiquitin seems to correlate with its reactivity towards the HA-UbBr2 probe with little to no reactivity towards HA-UbVME (Fig. 6B), whereas OTUB2 reacts with both Br2 and VME probes and does exhibit a more permissive cleavage profile including Lys48-, Lys63 (di/tetra-Ub)—and K6/K11 (di-Ub)-linkages (Figs. 5D, 6B and CߝG). The reason for the differential probe reactivity is not exactly understood, but clearly indicates subtle alterations within the catalytic cleft region between OTUB1 and OTUB2. In addition, structural elements other than the catalytic site must play a role as their ubiquitin chain linkage preference is also reflected by using di/tetra-ubiquitin substrates without electrophilic moieties for trapping the active site cysteine [18, 19, 35]. Crystallographic evidence suggested that the N-terminal α-helix of OTUB1 (Fig. 7, dark blue cylinder) [13, 14] that is absent in OTUB2 makes direct contact with the proximal ubiquitin and hence restricts its binding to an orientation presenting Lys48 towards the catalytic site (Figs. 4C and 7, red arrows). This restriction is not present in OTUB2, thereby potentially allowing a more permissive ubiquitin recognition mode [14, 15, 18]. OTU DUBs have been classified into different subgroups, in which OTUB1 belongs to enzymes with high selectivity for specific Ub-linkages (group I), whereas OTUB2 belongs to a set of enzymes with specificity to three of more linkage types (group III) [18]. OTUB1 and also DUBA N-terminal domains are posttranslationally modified with phosphate groups that influence their activity and/or substrate interaction [8, 36, 37]. The role of the N-terminal domain combined with some differences observed in within the catalytic cleft of OTUB1 and OTUB2 [19] could explain, at least in part, the observed differences in Ub-linkage cleavage specificity (Fig. 7). Also, it appears that other determinants, e.g. the α2α3 loop or more likely, yet to be identified interaction surfaces with the distal Ub, may be responsible for conferring chain specificity to OTUB1. Our results would be compatible with an auto-inhibitory function of the N-terminal OTUB1 helix.


The human otubain2-ubiquitin structure provides insights into the cleavage specificity of poly-ubiquitin-linkages.

Altun M, Walter TS, Kramer HB, Herr P, Iphöfer A, Boström J, David Y, Komsany A, Ternette N, Navon A, Stuart DI, Ren J, Kessler BM - PLoS ONE (2015)

OTUB2 has a broader cleavage profile than OTUB1.(A) Ubiquitin (Ub), Nedd8, ISG15 and SUMO1 were conjugated to the biotinylated peptide VKAKIQD (Ub26–32) as described in [19] and subjected to cleavage by recombinant OTUB2, UCH-L3 or crude cell lysate (- represents untreated control), followed by SDS-PAGE separation and analysis by streptavidin-HRP immunoblotting. (B) di-SUMO2/3 was incubated with DMSO, OTUB2, OTUB2delta, OTUB1 or crude lysate for the indicated times, followed by SDS-PAGE separation and analysis by immunoblotting. (C) Linear di-Ubiquitin (di-Ub) was incubated with OTUB2, UCH-L3 or DMSO for the indicated times, followed by SDS-PAGE separation and analysis by immunoblotting. (D) di-Ub substrates with the linkages Lys6, 11, 27, 29, 33, 48 or 63 were incubated with either wild type (Wt) or catalytically inactive C51S mutant (M) OTUB2 for four hours, followed by SDS-PAGE and immunoblotting analysis.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4295869&req=5

pone.0115344.g005: OTUB2 has a broader cleavage profile than OTUB1.(A) Ubiquitin (Ub), Nedd8, ISG15 and SUMO1 were conjugated to the biotinylated peptide VKAKIQD (Ub26–32) as described in [19] and subjected to cleavage by recombinant OTUB2, UCH-L3 or crude cell lysate (- represents untreated control), followed by SDS-PAGE separation and analysis by streptavidin-HRP immunoblotting. (B) di-SUMO2/3 was incubated with DMSO, OTUB2, OTUB2delta, OTUB1 or crude lysate for the indicated times, followed by SDS-PAGE separation and analysis by immunoblotting. (C) Linear di-Ubiquitin (di-Ub) was incubated with OTUB2, UCH-L3 or DMSO for the indicated times, followed by SDS-PAGE separation and analysis by immunoblotting. (D) di-Ub substrates with the linkages Lys6, 11, 27, 29, 33, 48 or 63 were incubated with either wild type (Wt) or catalytically inactive C51S mutant (M) OTUB2 for four hours, followed by SDS-PAGE and immunoblotting analysis.
Mentions: We have searched for evidence for regulation of OTUB2 enzymatic activity. As shown previously, OTUB2 cleaved a Ub-based peptide substrate harbouring an isopeptide bond (Fig. 5A). Interestingly, we also noted cross-reactivity towards cleaving a NEDD8-based peptide substrate (Fig. 5A), although this may be a substrate-specific trait [18]. OTUB2 did not show any activity towards the ISG15-based peptide substrate, SUMO1, 2 or 3 (Fig. 5A and B) nor linear di-Ub (Fig. 5C). In contrast to OTUB1 which has exclusive specificity towards Lys48-linked chains [34], OTUB2 cleaves a broader range of di-Ub linked by naturally occurring isopeptide linkages with a preference for Lys63 di-Ub (Fig. 5D), consistent with previous studies [18]. A short C-terminal truncation (1–229; OTUB2ΔC5) did not markedly affect activity (S1A and C Fig., [5]), and no post-translational modifications within the protein were detected (data not shown). OTUB1’s strict selectivity towards cleaving Lys48-linked poly-Ub chains is in part due to its N-terminal properties [14, 15, 18]. OTUB2 has a shorter N-terminal tail and therefore might lack this feature to control for cleavage specificity. To test this hypothesis, we prepared chimeric constructs where the N-terminal tails of OTUB1 and OTUB2 were swapped to create N-term OTUB1-OTUB2 (OTUB1–2) and N-term OTUB2-OTUB1 (OTUB2–1) recombinant proteins (Fig. 6A). The OTUB1 N-terminal tails (1–83) and OTUB2 (1–43) were designated such that the OTU domain was left intact (S2 Fig.). Interestingly, active site labeling with either Br2 or VME based ubiquitin probes indicated that the OTUB1 N-terminal tail affects labeling selectivity of OTUB2 (OTUB1–2) towards the VME probe (Fig. 6B). Furthermore, OTUB2 enzymatic activity was restricted due to the presence of the OTUB1 N-terminal tail, and OTUB1 activity was enhanced in the presence of the OTUB2 N-terminal tail (OTUB2–1, Fig. 6C, D and E). Consistent with this, we observed that the presence of the OTUB1-N-terminal tail on OTUB2 (OTUB1–2) influenced its selectivity to cleave Lys63-tetra-ubiquitin chains when wild type and chimera OTUB1&2 recombinant proteins were subjected to a tetra-ubiquitin cleavage assay (Fig. 6F and G). Notably, the exclusive selectivity of OTUB1 for Lys48-linked di/tetra-ubiquitin seems to correlate with its reactivity towards the HA-UbBr2 probe with little to no reactivity towards HA-UbVME (Fig. 6B), whereas OTUB2 reacts with both Br2 and VME probes and does exhibit a more permissive cleavage profile including Lys48-, Lys63 (di/tetra-Ub)—and K6/K11 (di-Ub)-linkages (Figs. 5D, 6B and CߝG). The reason for the differential probe reactivity is not exactly understood, but clearly indicates subtle alterations within the catalytic cleft region between OTUB1 and OTUB2. In addition, structural elements other than the catalytic site must play a role as their ubiquitin chain linkage preference is also reflected by using di/tetra-ubiquitin substrates without electrophilic moieties for trapping the active site cysteine [18, 19, 35]. Crystallographic evidence suggested that the N-terminal α-helix of OTUB1 (Fig. 7, dark blue cylinder) [13, 14] that is absent in OTUB2 makes direct contact with the proximal ubiquitin and hence restricts its binding to an orientation presenting Lys48 towards the catalytic site (Figs. 4C and 7, red arrows). This restriction is not present in OTUB2, thereby potentially allowing a more permissive ubiquitin recognition mode [14, 15, 18]. OTU DUBs have been classified into different subgroups, in which OTUB1 belongs to enzymes with high selectivity for specific Ub-linkages (group I), whereas OTUB2 belongs to a set of enzymes with specificity to three of more linkage types (group III) [18]. OTUB1 and also DUBA N-terminal domains are posttranslationally modified with phosphate groups that influence their activity and/or substrate interaction [8, 36, 37]. The role of the N-terminal domain combined with some differences observed in within the catalytic cleft of OTUB1 and OTUB2 [19] could explain, at least in part, the observed differences in Ub-linkage cleavage specificity (Fig. 7). Also, it appears that other determinants, e.g. the α2α3 loop or more likely, yet to be identified interaction surfaces with the distal Ub, may be responsible for conferring chain specificity to OTUB1. Our results would be compatible with an auto-inhibitory function of the N-terminal OTUB1 helix.

Bottom Line: Here we report the crystal structure of human otubain 2 (OTUB2) in complex with a ubiquitin-based covalent inhibitor, Ub-Br2.The ubiquitin binding mode is oriented differently to how viral otubains (vOTUs) bind ubiquitin/ISG15, and more similar to yeast and mammalian OTUs.In contrast to OTUB1 which has exclusive specificity towards Lys48 poly-ubiquitin chains, OTUB2 cleaves different poly-Ub linked chains.

View Article: PubMed Central - PubMed

Affiliation: Target Discovery Institute, Nuffield Department of Medicine, Roosevelt Drive, University of Oxford, Oxford, OX3 7FZ, United Kingdom; Science for Life Laboratory, Division of Translational Medicine and Chemical Biology, Department of Medical Biochemistry and Biophysics, Karolinska Institutet, SE-171 21 Stockholm, Sweden.

ABSTRACT
Ovarian tumor domain containing proteases cleave ubiquitin (Ub) and ubiquitin-like polypeptides from proteins. Here we report the crystal structure of human otubain 2 (OTUB2) in complex with a ubiquitin-based covalent inhibitor, Ub-Br2. The ubiquitin binding mode is oriented differently to how viral otubains (vOTUs) bind ubiquitin/ISG15, and more similar to yeast and mammalian OTUs. In contrast to OTUB1 which has exclusive specificity towards Lys48 poly-ubiquitin chains, OTUB2 cleaves different poly-Ub linked chains. N-terminal tail swapping experiments between OTUB1 and OTUB2 revealed how the N-terminal structural motifs in OTUB1 contribute to modulating enzyme activity and Ub-chain selectivity, a trait not observed in OTUB2, supporting the notion that OTUB2 may affect a different spectrum of substrates in Ub-dependent pathways.

Show MeSH
Related in: MedlinePlus